Bringing Conjugated Polymers and Oxide ... - ACS Publications

Article pubs.acs.org/JPCC

Bringing Conjugated Polymers and Oxide Nanoarchitectures into Intimate Contact: Light-Induced Electrodeposition of Polypyrrole and Polyaniline on Nanoporous WO3 or TiO2 Nanotube Array Csaba Janáky,*,†,‡ Norma R. de Tacconi,† Wilaiwan Chanmanee,† and Krishnan Rajeshwar*,† †

Department of Chemistry and Biochemistry, University of Texas at Arlington, Arlington, Texas 76019, United States Department of Physical Chemistry and Materials Science, University of Szeged, Szeged, H6720, Hungary

‡

S Supporting Information *

ABSTRACT: This proof-of-concept study focuses on the photocatalytic electrodeposition of two conducting polymers, namely, polyaniline (PANI) and polypyrrole (PPy), in two different nanostructured inorganic semiconductor host matrices, namely, nanoporous tungsten trioxide and nanotubular titanium dioxide. Oxide semiconductor (WO3 and TiO2) films were initially electrosynthesized on tungsten and titanium foils, respectively, by anodization at different voltages in fluoride-containing aqueous media. The conjugated polymer was electrografted onto the entire surface of the photoexcited oxide semiconductor matrix using potentiostatic and potentiodynamic deposition methods. The crucial role of initial photoelectrochemical deposition, preceding the electrochemical polymerization step, was demonstrated. The photoelectrodeposited and electrodeposited hybrid samples were compared from both morphological and electrochemical perspectives. Importantly, through application of the methodology presented in this article, deposition of electroactive polymers can be achieved homogenously, on both macroscale and nanoscale dimensions. The morphology and structural properties of these assemblies were evaluated by FE-SEM, ATR FT-IR, and Raman spectroscopy, whereas their electroactivity was characterized by cyclic voltammetry.

■

INTRODUCTION

Such hybrid materials can be obtained by different synthetic routes ranging from simple mechanical mixing of the components, through chemical or electrochemical polymerization of the monomer in the presence of the inorganic nanoparticles, to simultaneous electrochemical codeposition of the polymer and the oxide (nano)particles. All of these methods possess similar drawbacks and limitations, the three most important being: (i) potential aggregation of the inorganic particles, resulting in a relatively small area of the p/n junction; (ii) uncontrolled, random distribution of the particles within the polymeric matrix; and finally (iii) lack of electrical contact between the inorganic material with the supporting electrode. Therefore, the use of composites prepared thus as electrodes or in electronics (which is the case in most of the applications that have been considered) is hindered by the fact that only the

Hybrid materials based on organic conducting polymers (CPs) and inorganic nanostructures have been at the leading edge of research and development.1−4 A wide range of composites has been realized, including the hybridization of metal nanoparticles, carbon nanostructures, and metal chalcogenides (specifically quantum dots) with CPs.3,4 Oxide semiconductors are remarkably attractive materials in this regard because a wide range of properties can be combined with complementary features of the organic counterpart. Therefore, p/n junctions were formed by combining various polymers (dominantly thiophene derivatives) with TiO2, WO3, or ZnO and deployed in solar cells,5−8 electrochromic devices,9 or for photocatalysis.10 Incorporation of MoO311 and Fe3O412 into CPs has led to hybrid materials with superior sensing, magnetic, and catalytic properties. Composite materials with enhanced charge storage capacity were prepared by embedding V2O5, RuO2, and MnO2 into the matrix of CPs such as polypyrrole,13 polyaniline (PANI),14 and PEDOT.15 © 2012 American Chemical Society

Received: May 28, 2012 Revised: July 5, 2012 Published: July 11, 2012 19145

dx.doi.org/10.1021/jp305181h | J. Phys. Chem. C 2012, 116, 19145−19155

The Journal of Physical Chemistry C

Article

PPV), poly(3,4 ethylenedioxithiophene) (PEDOT), or polythiophene, respectively, in the oxide host. Electrodeposition is a particularly versatile method in this regard because it relies on exploitation of the intrinsic electroactivity of the monomer. The most important possible advantage of this approach is that the CPs can be directly electrogenerated on the surface of the inorganic nanostructure, which acts as a working electrode. Therefore, this method has been employed to deposit CPs on different nanostructured semiconductor matrices such as TiO2,38−42 ZnO,43 Fe2O3,44 or WO3.45 Moreover, because of the advancement of the knowledge on inorganic materials, there is a fairly large number of different anodized nanostructures.19 Therefore, there is a potentially unlimited number of combinations with various CPs. Nonetheless, wide bandgap semiconductors (including TiO2) are handicapped by their low electrical conductance, especially in the dark. This results in inhomogeneous deposition of the polymer, either on the top of the substrate, or at the supporting metal electrode interface due to cracks or other defects in the oxide coverage. Importantly, because of inadequate filling of the nanopores the area of the p/n junction is limited, thus deleteriously affecting the performance of the hybrid in different applications. To underline the role of electroactivity of the oxide semiconductor, Figure 1 presents a comparison of cyclic

conductive polymer matrix has direct electrical connection to the electrode substrate. Nanostructuring of one or both components of the hybrid assembly is very attractive from both scientific and technological aspects. In fact, the role of morphology and nanoscale structure within organic/inorganic assemblies has been recently highlighted.16,17 Organized nanostructured frameworks of inorganic semiconductors with well-defined morphologies such as TiO2 nanotube array (NTA)18,19 or nanoporous WO320,21 are particularly eminent candidates to overcome the problems listed above. By infiltrating CPs into such nanoarchitectures, hybrid assemblies with ordered structure, large areas of organic/inorganic junctions, and distinct morphologies, can be obtained. In such nanostructured configurations, both components will have better electrical contact with the underlying electrode. For the preparation of the inorganic part of these hybrid materials, anodization is indeed an attractive method because the resulting oxides fulfill all of the enumerated requirements. Most importantly, they have an ordered nanoscale structure, which greatly enhances semiconductor behavior, particularly improved charge carrier collection due to 1D vectorial e− transport.22−24 Moreover, through simple adjustment of the parameters of the synthetic procedure, several key properties of the nanotubes (length, diameter, wall thickness, conductance, etc.) can be tuned for targeted applications. Because of these favorable properties, TiO2 NTAs are undoubtedly one of the most intensively studied structures within this class of materials. Accordingly, a wide scale of utilization possibilities have been demonstrated already, such as photocatalysis, electrochromism,19 solar cells,22−25 or drug delivery.26 Similarly, nanoporous WO3 obtained via anodization of W-foil showed superior efficiency compared with counterparts synthesized by other methods, in terms of photoelectrochemical behavior, electrochromism, sensing, electrocatalyic activity, or water splitting ability.27−29 Realization of hybrid materials based on the above organized oxide nanostructures and CPs usually involves spin coating of the solution of the polymer in organic solvents, followed by thermal treatment. However, this approach has limited scope due to simple physical reasons, most importantly because colloidal size polymeric macromolecules are too big to penetrate into the pores of a nanostructured matrix (especially if it is already partially loaded), resulting in insufficient pore filling. The net result is that the organic component resides dominantly at the upper areas of the NTA. Interfacial contact between the inorganic oxide and the infiltrated polymer is rather poor because there is no driving force that would ensure intimate contact between the organic and the inorganic components. Recent studies have underlined the crucial role of the nature of the organic/inorganic interface for different applications.30,31 Consequently, there are only a few examples for successful application of hybrid assemblies, dominantly as solar cells with32 or without33,34 dye sensitization. Importantly by solving the morphological and structural issues listed above, enhanced performance may be expected for organic/inorganic hybrids in a wide range of applications. In an effort to overcome the above-mentioned obstacles and to achieve homogeneous infiltration of the CP component, different in situ approaches using the corresponding monomer precursor have been deployed. Chemical,35 thermal,36 or UV photopolymerization37 was utilized to synthesize poly(1methoxy-4-(2-ethylhexyloxy)-pphenylene-vinylene) (MEH-

Figure 1. Cyclic voltammograms of nanoporous WO3 and TiO2 NTA electrodes at a potential sweep rate of 100 mV s−1 in 0.5 M H2SO4 and in 0.025 M SDS. The inset shows the current values registered at E = 1.0 V.

voltammograms for TiO2 NTA and nanoporous WO3 in acidic (H2SO4) and neutral (sodium dodecyl sulfate, SDS) media. Note that all samples are electroactive in the negative potential regime; however, the magnitude of this activity is different for each system. More importantly, as can be clearly seen in the inset in Figure 1, the electroactivity of the materials is very low at the anodic end of the cycles, except in the case of the WO3/ H2SO4 interface. This enhanced electroactivity of WO3 in acidic media was rationalized by the formation of tungsten bronze (HxWO3), assisted by H+ uptake/release as charge compensation during the (W6+/W5+) redox process.27 Because of the low electroactivity of the oxide semiconductor working electrode, effective and homogeneous anodic electrodeposition can only be achieved under very carefully designed 19146



Article

and used immediately. For PANI and polypyrrole growth, aniline (EM Science, 99.5%) and pyrrole (Aldrich, 98.0%) monomers were freshly distilled under vacuum before use. Sulfuric acid (Alfa Aesar, 98.0%), sodium dodecyl sulfate (SDS, Sigma-Aldrich, 98.5%), and sodium-sulfate (J.T. Baker, 99.0%) were used as received. Anodic Growth of Nanoporous WO3 and TiO2 NTA. Nanoporous WO3 and nanotubular TiO2 films were grown in a two-electrode electrochemical cell using a large Pt foil as counter electrode and the respective metal foil as the working electrode. Aqueous NaF was used as the electrolyte (0.15 M) in the case of W, whereas Ti foil was anodized in solutions containing 0.36 M NH4F in poly(ethylene glycol) containing a 20% v/v of deionized water. During anodization, the metal foil was pressed between a set of O rings in the electrochemical cell, leaving 0.78 cm2 exposed to the electrolyte, and the electric contact was located on the backside of the sample. Anodization employed a 420X Power Supply (The Electrosynthesis Company, Lancaster, NY). The voltage was held at a preselected level (60 V in the case of WO3 and 20 V for TiO2, respectively) for 3 h. After the oxide film was grown, the anodized W/Ti foil was removed from the O-ring assembly, carefully washed by immersion in deionized water, and then dried in a N2 stream. Before use as substrates for aniline/ pyrrole polymerization, the W/WO3 and Ti/TiO2 electrodes were annealed at 450 °C (under air atmosphere) for 3 h at a heating rate of 20 °C/min (Fisher Scientific, model 650-14 Isotemp Programmable Muffle Furnace) and allowed to cool gradually back to the ambient temperature. Other details of anodic growth are given elsewhere.21 Electrosynthesis of the Hybrid Materials. A similar experimental setup was used as that during anodization. The working electrode was the previously prepared nanoporous WO3 or the nanotubular TiO2 in all cases. The reference electrode was Ag/AgCl/satd. KCl (Microelectrode, Bedford, NH); all potentials in this article are given with respect to this reference. Pt coil was utilized as the counter electrode. The parameters of the performed experiments are summarized in Table 1. Each electrochemical deposition was performed both

experimental conditions, both in terms of polymerization electrolyte and procedure. A noteworthy method was reported recently involving several very short steps, employing high current densities and electrode potentials, and was successfully deployed to fill homogenously either the TiO2 nanotubes or the interannular space among the nanotubes with polypyrrole.46,47 This method was developed based on an analogy to metal deposition on nanoporous alumina template.48 However, the electroactivity of such deposited polymers can be problematic due to potential overoxidation under these rather harsh conditions. As another example, utilization of specially designed monomers with functional groups, which can be covalently bonded to the inorganic template, can also enhance homogeneity of the hybrid morphology and improve its interfacial properties.49 Finally, uniform deposition of PANI on nanoporous WO345 was accomplished by recognizing the similarity in their ion exchange properties. That is, both of them are electron and proton conductors resulting in reasonable WO3 electroactivity, even in the potential regime required for electropolymerization of PANI (see Figure 1). Because the predominant limitation of these wide bandgap nanostructured oxide matrices is their low electroactivity in the potential regime where CPs are usually electrogenerated and at the same time electronic carriers can be generated in these matrices by photoexcitation, our aim in this study was to study the effect of illumination on the electrochemical deposition of the CP component. We note literature precedent for this notion.50−56 Some monomers can also be photopolymerized by UV light, even without the presence of a semiconductor57,58 but usually in the presence of an electron scavenger and often in the presence of a homogeneous photocatalyst. Electropolymerization of EDOT was also performed under illumination, and thick PEDOT and PPy layers were deposited on f lat Nb2O5 and Ta 2O5 substrates, respectively, for solid capacitor applications.59,60 Notably, to the best of our knowledge, there is no record in the literature of employing illumination to enhance the electrodeposition of CPs on nanostructured oxide hosts, nor are there attempts to improve the interfacial properties of resulting hybrid assemblies based on such oxides. In this article, we present the dramatic effect of illumination on the electrodeposition of two CP candidates in nanostructured oxide semiconductor host frameworks. Moreover, we uncover important mechanistic aspects and clarify the relative contribution of photoelectrochemical deposition and electrochemical polymerization to the overall process.

Table 1. Parameters of the Electropolymerization Procedure

■

EXPERIMENTAL SECTION Chemicals and Materials. All chemicals were from commercial sources and were of the highest purity available. Deionized water (18 MΩ cm) was used in all cases for making solutions. Sodium fluoride (Alfa Aesar, 98.0%), poly(ethylene glycol) (Mn ca. 400, Aldrich), and ammonium fluoride (Alfa Aesar, 98.0%) were used as received. Tungsten (Alfa Aesar, 0.25 mm thick, 99.95%) and titanium (Alfa Aesar, 0.25 mm thick, 99.95%) foils were used as the substrate for oxide film growth. Before use, the foil was cut (1.25 cm × 1.25 cm), mechanically polished to mirror finish using silicon carbide sandpaper of successively finer roughness (220, 400, 600, 1000, 1500, and 2000 grit), and cleaned in three 5 min steps in ultrasonicated acetone, 2-propanol, and finally ultrapure water. Subsequently, the substrate was dried in an ultrapure N2 stream

hybrid material

polymerization solution

polymerization method

WO3/ PANI WO3/ PPy TiO2/ PANI

0.2 M aniline, 0.5 M H2SO4 0.1 M pyrrole, 25 mM SDS 0.2 M aniline, 0.5 M H2SO4

potentiodynamic (−0.2 V − 1.1 V), potentiostatic (E = 0.85 V) potentiodynamic (−0.4 V − 1.1 V), potentiostatic (E = 0.90 V) potentiodynamic (−0.2 V − 1.1/1.45/2.2 V), potentiostatic (E = 1.45 V)

in the dark and under illumination, whereas all other circumstances were kept identical. The effect of the applied potential window was also studied, as can be seen in Figure 8. For photoelectrochemical deposition, the light source was a 150 W xenon arc lamp (Oriel, Stratford, CT), and the radiation source was placed 8 cm away from the working electrode surface. The illuminated area was 0.13 cm2, whereas the overall electrode surface exposed to solution was 0.39 cm2. Selective illumination of the working electrode was performed by using a mask on top of it with a transparent pattern (ACS, see Figure S2 of the Supporting Information). (See Figures 4 and 7 below) 19147



Article

Figure 2. Comparison of the electrodeposition of polyaniline on nanoporous WO3 with and without illumination (150 W Xe-arc lamp). (A) Potentiodynamic cycles were recorded between −0.2 and 1.1 V in a solution of 0.2 M aniline in 0.5 M H2SO4. (B) Cumulative charge flow during the subsequent redox cycles in panel A. (C) Enlargement of the first half-cycles depicted in panel A.

■

RESULTS AND DISCUSSION Photoelectrodeposition of PANI on WO3. The effect of illumination was first studied employing a hybrid material described recently,45 namely, the WO3/PANI as an example. As we previously presented, because of its H+ exchanging (bronze formation) behavior, WO3 is sufficiently electroactive for dark electrodeposition on it in strongly acidic media (Figure 1). However, if the electrochemical polymerization is performed under illumination but otherwise identical circumstances, then the overall pattern of current−potential curves registered during potentiodynamic cycling is dramatically different. As can be seen in Figure 2A, rapid growth of the polymer (as diagnosed from the current flow) is striking, although the shapes of the individual voltammograms (cycles) are indeed similar to those obtained in the dark or during electrodeposition on metal electrodes. By integrating the current registered during the subsequent polymerization cycles, more quantitative data may be gleaned on the polymerization process (and the amount of the generated polymer) in the two cases. The cumulative charges are depicted in Figure 2B, where the large difference between the two cases is distinctly visible. Careful examination of the first half-cycles of the polymerizations sheds further light on mechanistic aspects of the oligomer/polymer formation. The curves shown in an enlarged view in Figure 2C are dissimilar in many aspects. For the WO3 sample without illumination, very low currents are detected throughout the entire potential window. This trend is typical for CP deposition on semiconductor electrodes because during the first cycle there is only oligomerization at the positive end of the potential range. Growth of the polymer phase proceeds during successive cycles, when the template is already occasionally covered by oligomeric/polymeric nucleating seeds on which the polymer can grow further at even lower potentials. In other words, PANI has an autocatalytic effect on the electropolymerization of aniline.61,62 The complexity of the current response under illumination is very distinct. At the beginning of the curve, anodic photocurrent develops at ∼E = −0.05 V, which is consistent with the n-type semiconductor behavior of WO3. The photocurrents arise mainly from the photooxidation of adsorbed water, electrolyte sulfate ions, or more likely the aniline monomer moieties present in the solution. Interestingly the photocurrent does not increase further with increasing bias as expected: in contrast, it starts to decrease from E = 0.10 V onward. This anomalous behavior can be attributed to the photocatalytic

According to our previous study, for the deposition of PANI, a polymerization medium consisting of 0.2 M aniline monomer and 0.5 M aqueous sulfuric acid was chosen.45 The solution composition for polypyrrole deposition was adopted from a preceding work. 47 Optimization of the electrochemical polymerization conditions was partially carried out in a previous study (on WO3/PANI),45 and further developed herein. Finally, polymerization of aniline and pyrrole was carried out by simple but carefully optimized potential cycling protocols between different potential ranges adjusted to each system. (See Table 1.) On the basis of the data obtained during the potentiodynamic deposition, potentiostatic polymerization was also performed. For further voltammetric studies, the solution was changed after polymerization to a monomer-free 0.5 M H2SO4 and 0.1 M Na2SO4 electrolyte. All experiments were carried out under ambient conditions at room temperature (25 ± 2 °C). Characterization Methodology. The morphology of the bare nanostructured oxides (WO3, TiO2) and the respective hybrid samples was studied using a Hitachi S-5000H fieldemission scanning electron microscope (FE-SEM) at an accelerating voltage of 15 and 20 kV. Images were taken at different magnifications between 10k and 200k. The molecular structure of the deposited CPs was investigated by vibrational spectroscopy. FT−IR spectra were recorded using a Shimadzu IRPrestige-21 Fourier transform infrared spectrometer equipped with a diamond attenuated total reflectance (ATR) accessory. All infrared spectra were recorded between 4000 and 380 cm−1 at 4 cm−1 optical resolution by averaging 64 interferograms. Raman spectra were recorded with a Horiba Jobin Yvon LabRam ARAMIS instrument (P ≤ 300 mW) using an excitation wavelength of 473 nm and 1800 line/mm grating. In all of the cases, the slit width was 10 μm, and 10 scans were accumulated for each spectrum. A Nikon Eclipse optical microscope equipped with a camera was employed to acquire pictures on the hybrid samples. All electrochemical measurements were performed on a CHI Electrochemical Workstation 440A instrument in a classical one-compartment, threeelectrode electrochemical cell. Cyclic voltammograms of the WO3/PANI and TiO2/PANI hybrid samples were registered in 0.5 M H2SO4 solutions, whereas for WO3/PPy 0.1 M Na2SO4, supporting electrolyte was used at four different potential sweep rates between 10 and 100 mV/s. 19148



Article

Figure 3. Photochronoamperometric curves for nanoporous WO3, recorded in (A) 0.5 M H2SO4 and 0.2 M aniline and (B) 0.025 M SDS and 0.1 M pyrrole. The curves recorded in the absence of monomers are also shown for comparison at E = 0.85 and 0.9 V, respectively.

Figure 4. Comparison of electrodeposition of polypyrrole on nanoporous WO3 with and without illumination. (A) Potentiodynamic cycles recorded between −0.2 to 1.1 and 1.25 V in the dark and under illumination, respectively, in a solution of 0.1 M pyrrole in 0.025 M SDS. (B) Enlargement of the first half-cycles depicted in panel A. The inset shows an optical image of a sample photoelectrodeposited under selective illumination through a mask. (See Figure S2 of the Supporting Information.)

polymerization of aniline, more precisely to the optical (and electric) shielding effect of the photoelectrochemically deposited PANI phase. As a first step, aniline monomers are oxidized by photogenerated holes on the WO3 surface. Then, the radical cation reacts with another radical cation to form dimers that can react further, with either a monomer or an oligomer radical cation.63 As the oligomers become progressively insoluble, they deposit on the surface of the nanostructured electrode substrate. When the potential is increased further, at a certain threshold value (E = 0.80 V) electrochemical growth of the polymer starts to take place and proceeds rapidly. The striking difference between the two curves in this potential range can be explained by the fact that under illumination oligomer/polymer formation already took place at lower potentials due to photoelectrochemical deposition, thus obviating the need for electrochemical oxidation of the aniline monomer, which requires notably higher potentials. It is worth noting the similarity in the shape of the anodic ends of the cycles for the first cycle under

illumination and the later cycles registered in the dark. This is further indication of the fact that electrochemical growth of the polymer is similar in the two cases, just the formation of the seed layer is different: photoelectrochemically deposited under illumination and electrodeposited in the dark. Data obtained during potentiodynamic experiments suggested that PANI can be photocatalytically polymerized even below the potentials where electrochemical polymerization occurs. However, the external bias potential plays a key role even in this process because it facilitates the separation of the photogenerated charge carriers (by draining the photoelectrons to the back contact and hampering recombination). In this manner, more holes can react with the aniline monomers and oligomers to initiate/facilitate their polymerization through the formation of radical cations. Chronoamperometric curves under illumination and registered at E = 0.85 V are presented in Figure 3A, both in the presence and in the absence of aniline monomer. The selection of this potential value was motivated by the fact that at this 19149



Article

Figure 5. SEM images of the bare WO3 surface synthesized at E = 60 V (A,B) and of the corresponding WO3/PPy hybrid samples (synthesized using five cycles with (C,D) and without (E,F) illumination) at 50k and 100k magnification.

of higher currents (electroactivity) in the series of potentiodynamic cycles (Figure 4A). Consider the first half-cycles enlarged in Figure 4B; although the overall profile is very similar to that presented for WO3/ aniline (Figure 2C), the effect of illumination is even more expressed. The one difference is that photocurrents can be detected from the very beginning of the potential window due to the very good hole scavenging activity of pyrrole and oligopyrroles formed because of UV illumination. Later, a slight decrease in the photocurrent can be noticed due to shielding by the photoelectrochemically deposited PPy, whereas from E = 0.7 V a rapid upturn of the current can be seen, as an indication of the electropolymerization of pyrrole. Notably, under illumination, deposition of PPy takes place only at the illuminated region of the working electrode surface, resulting in homogeneous coverage. As further evidence of the selectivity of the process, by employing a mask, just certain areas of the WO3 electrode were illuminated (see the Experimental Section and Figure S2 of the Supporting Information), whereas the whole electrode was exposed to the solution. The result is depicted as an inset of Figure 4A; clearly, polymerization is limited to the areas exposed to light, and thus the pattern of the deposited polymer is identical to the pattern of illumination. Finally, by comparing Figures 2A and 4A, we can conclude that deposition of PPy is less favorable on WO3 than that of PANI; however, the illumination plays a very similar role in both cases. To compare the nanoscale morphological features of the hybrid assemblies and to gain further insight into the effect of illumination, SEM images were taken at various magnifications, with both the bare WO3 nanostructure and the hybrid samples. In Figure 5A,B, SEM images of the bare oxide samples are presented. As can be clearly seen, WO3 has a very porous structure with high surface area, reflecting effective overetching of the tungsten surface by the fluoride species during anodization. Figure 5C,D shows representative SEM images for the hybrid samples, obtained for the deposition of PPy with five potentiodynamic cycles on the WO3 surface under illumination. The most noticeable differences compared with the bare oxide support are (i) the opacity seen for the WO3 surface on the whole sample, which can be generally observed for organic polymers; (ii) the expansion of the grain size to the

potential polymerization cannot be initiated in the dark (monomer oxidation does not take place), but growth of the polymer on a predeposited oligomeric/polymeric film can indeed occur. (See Figure 2.) As can be distinctly seen in Figure 3A, little current response can be obtained without illumination, either in the absence or in the presence of aniline. (Note that the curve registered in the presence of the monomer is vertically shifted, for better visualization.) Under illumination, WO3 shows the regular photocurrent transient profile in the absence of aniline, whereas a rather interesting shape can be observed when the monomer is present in the solution. As soon as the light is turned on, the current starts to develop rapidly, but instead of reaching a maximum value, it increases continuously. This pattern is related to the previously described steps of the deposition, namely, that first only photoelectrochemical deposition takes place, but it is immediately followed by electrochemical growth of the polymer. Importantly, after turning off the light, there is only a small drop in the current, indicating that in this stage of the process electropolymerization has the dominant contribution compared with the photocatalytic process. WO3/Polypyrrole Hybrid. Although WO3/PANI hybrid previously presented is a good system to understand the basic aspects of role of illumination, the real challenge is to exploit this procedure for the preparation of hybrid materials, which cannot be obtained by other methods with the desired properties. As we showed in Figure 1, WO3 has much smaller electroactivity in nonacidic medium. However, several polymers (e.g., polypyrrole) cannot be generated in their conducting form in strongly acidic electrolyte. Motivated by these facts, we selected the WO3/PPy system as a candidate to study the effectiveness of our approach. As shown in Figure 4A, without illuminating the WO3 electrode, the anodic growth end of the potential window had to be shifted toward more positive potentials to obtain even a very slight amount of polymer. Moreover, visual examination reveals the distribution of the polymer to be rather inhomogeneous, with polymeric spots on the top of the WO3 electrode. In contrast, under illumination, drastically higher currents can be detected, and the formation of conducting PPy can be deduced from the gradual development 19150



Article

Figure 6. Raman (A) and FT-IR (B) spectra of nanoporous WO3, WO3/PANI, and WO3/PPy hybrid samples prepared by five potentiodynamic cycles under illumination. The insets show enlarged images of the areas characteristic for the two polymers.

Figure 7. Comparison of the electrodeposition of polyaniline on TiO2 NTA with and without illumination (150 W Xe-arc lamp). (A) Potentiodynamic cycles were recorded between −0.2 and 2.2 V, in a solution of 0.2 M aniline in 0.5 M H2SO4. (B) Enlargement of the first halfcycles depicted in panel A. (C) Cumulative charge flow during the subsequent redox cycles in panel A. (D) Optical image of selectively photoelectrodeposited PANI on the TiO2 NTA substrate.

polymeric film is deposited on it. On the basis of these observations, the most prominent difference between the two hybrid samples (prepared with and without illumination) is the contact area of the organic/inorganic junction, which is practically equal to the surface area of the inorganic semiconductor matrix in the case of the illuminated sample, whereas it is almost negligible for its counterpart synthesized in the dark. To confirm the molecular identity of the polymers photoelectrodeposited in the nanoporous WO3 framework, we carried out vibrational spectroscopic studies. FT-IR and Raman spectroscopies are rather useful for this purpose because WO3, PPy, and PANI all have characteristic bands. Moreover, beyond simple identification of hybrid components, these complementary techniques can also indicate if any permanent overoxidation of the polymer had occurred during electrodeposition. Figure 6A contains Raman spectra of a WO3 sample and the corresponding photoelectrodeposited WO3/PANI and WO3/PPy hybrid samples. Existence of vibrational modes64 at 133, 272, 325, 715, and 807 cm−1 confirms the formation of

parent situation in Figure 5A,B due to the homogeneous, skinlike coverage by the deposited polypyrrole; and (iii) the decrease in the space between the grains (the pore size). Closer inspection of the hybrid structure reveals a ∼50 nm average difference in the diameter between the neat WO3 grains and the hybrid sample, which suggests that polymer photoelectrodeposition results in a ∼25 nm thick PPy film on the surface. Under illumination, PPy is seen to deposit in a very similar fashion on WO3 as PANI in the dark,45 resulting in a comparable coating that covers the entire surface of the nanostructured semiconductor, as seen from both qualitative and quantitative analysis of the images. In contrast, images taken at the few spots where PPy was electrodeposited without illumination show completely different morphological attributes (Figure 5E,F). The two most striking alterations are the following: (i) instead of forming a thin film coverage on the nanoporous matrix, large globular polymeric units can be observed, randomly distributed on the top of the substrate and (ii) the WO3 grains have exactly the same average size as that measured for the bare oxide sample, which means that no 19151



Article

Scheme 1. Illustration of the Two Steps of the Photo-Electrodeposition of PANI on TiO2 NTA

Figure 8. Effect of the positive end expansion of the potential window on the polymerization of aniline on TiO2 NTA (five cycles) under illumination in 0.2 M aniline, 0.5 M H2SO4 solution, at a potential scan rate of 100 mV s−1.

−0.2 and +0.5 V) is larger even after one potentiodynamic cycle than that of five corresponding cycles in dark. The difference in the overall charge passed through during the subsequent cycles is also striking. As presented in Figure 7C, the slope of the line registered under illumination is more than six times larger than its counterpart obtained without illumination. Comparison of the first half-cycles shows the significant effect of light for this system as well. Photocurrent develops rapidly starting at E = 0.1 V, whereas after reaching its maximum, it starts to decrease owing to optical shielding by the photodeposited PANI (Figure 7B). By increasing further the bias potential, electropolymerization starts to contribute to the current, resulting in its increase at the positive end of the curve (Figure 7B). Hybrid samples obtained by illumination of selected areas of the TiO2 NTA working electrode provide visual proof of light-induced electrodeposition for PANI, as shown in Figure 7D. To shed further light on the contribution of photocatalytic and electrochemical polymerization to the formation of PANI (Scheme 1), we studied the role of the positive end of the potential window. More precisely, potentiodynamic cycles were run up to 1.1, 1.45, and 2.2 V, whereas other parameters remained unchanged. By appropriately selecting the applied potential regimes, the contribution of electropolymerization can be separated, which gives information on mechanistic aspects, and also the shape of the curves can be a good indication of the electroactivity of the polymers, obtained under different circumstances. As can be seen in Figure 8, a number of significant differences can be revealed. The most obvious alteration in the series of figures is probably the substantial increase in the magnitude of currents. As the potential window is extended, the contribution of electrochemical polymerization

monoclinic WO3 during anodization and subsequent heat treatment. Beyond the appearance of the bands originating from the support oxide framework, several new bands can be observed for the hybrid samples in Figure 6A. All characteristic bands can be assigned to the corresponding polymers. (See the assignments in the Supporting Information.)65−68 As for the PANI-containing hybrid, the overall pattern and the position of the characteristic bands reflects that the deposited PANI is in its emeraldine form, and thus no irreversible formation of pernigraniline had occurred during the polymerization. Similar conclusions can be drawn based on the infrared spectra, depicted in Figure 6B: (i) The broad absorption between 1000 and 600 cm−1 is characteristic of the different W−O stretching modes in the WO3 crystal lattice. (ii) Identification of the different additional vibrational bands on the spectra of the hybrids gave further confirmation for the formation of the corresponding CPs. (iii) Importantly, only a ν(CO) band of moderate amplitude can be observed around 1710 cm−1, which indicates a relatively small amount of overoxidized PPy. In the case of significant overoxidation, this band would have been much more expressed.69 Hybrids Based on TiO2 NTAs. Prompted by the data presented above for WO3, our synthesis concept was extended to TiO2 NTAs because hybrid materials based on this latter semiconductor may attract significant interest also from a utilization perspective. Very similar experiments were carried out with the TiO2/PANI system. To obtain PANI on a TiO2 NTA working electrode by electrodeposition, very high potentials have to be applied as depicted in Figure 7A. Similar to the examples previously presented, a massive increase in the currents can be observed under illumination. Note that the electroactivity related to the deposited polymer (between E = 19152



Article

Figure 9. SEM images of bare TiO2 NTA synthesized at E = 20 V (A,B) and of the corresponding TiO2/PANI hybrid samples (synthesized using five cycles with (C,D) and without (E,F) illumination) at 50k and 100k magnification.

Figure 10. Raman (A) and FT-IR (B) spectra of TiO2 NTA and the TiO2/PANI hybrid samples prepared by five potentiodynamic cycles under illumination. The insets show enlarged images of the areas characteristic for polyaniline.

is more and more expressed (compared with Figure 8A, where only photoelectrochemical polymerization occurs), first just causing a moderate enlargement in the registered currents (Figure 8B), whereas in the last case, the dominance of electropolymerization can be deduced (Figure 8C). This change is also reflected in the dissimilar shape of the curves. As for the anodic end of the curves (where electropolymerization generally occurs), interesting trends can be observed. For the narrowest potential window (Figure 8A), a gradual decrease is visible in the photocurrent in the order of cycles, which is dominantly related to the optical shielding effect of the photoelectrochemically deposited PANI. Oppositely, in the case of the widest potential range, a continuous increase can be noticed in the series of cycles due to the currents arising from electropolymerization. In the case of the middle potential window (Figure 8B), a mixed trend can be seen as a result of the competition between the decreasing photocurrent and the increasing polymerization current. Moreover, note the change in the profile of the curves between −0.2 and +0.6 V, which is related to redox transformation of the generated PANI. As can be seen in Figure 8A, the photodeposited polymer has a less-developed redox activity, the anodic peak around E = 0.3 V is less expressed, whereas only one cathodic peak can be seen

instead of the usually obtained double wave. This inferior electroactivity is in agreement with previous studies related to the photocatalytic deposition of CPs.52 In contrast, a welldefined, quasi-reversible redox behavior can be seen in the two other cases, where electropolymerization has an important contribution to polymer generation. This is especially true for Figure 8C, where the voltammograms mirror their counterparts on gold electrode. Moreover, the good electroactivity of PANI in the hybrid configuration was confirmed even by cyclic voltammetric studies, performed in monomer-free solution (Figure S1 of the Supporting Information). Figure 9 contains representative SEM data for both the initial TiO2 NTA surface (Figure 9A,B) and the TiO2/PANI hybrid synthesized under illumination (C,D) and in the dark (E,F). The bare nanotubes exhibit the typical properties for NTAs synthesized under these conditions, in terms of both average tube diameter (85 nm) and wall thickness (15 nm). The morphology of the photoelectrochemically prepared hybrid sample is strikingly different (Figure 9C,D). As can be seen from the expansion of wall thicknesses and from the disappearance of nanoscopic voids in between the tubes, PANI is deposited all over the illuminated area of the sample. It should also be noticed that although the nanotubes are 19153



■

completely covered by the polymeric skin, further growth of the polymer is preferred among the tubes, not inside them (Scheme 1).47 The hybrid sample obtained in the absence of illumination exhibits a morphology similar to that seen that with the WO3/PPy system. PANI is randomly, occasionally deposited on the TiO2 NTA, forming globular particles on the top of the substrate (note the similarities with precedent literature41,42). The individual tubes are not covered, nor is the space among the tubes filled with the conjugated polymer. Completely identical wall thicknesses can be observed with the parent (bare TiO2 NTA) situation. By comparing the two hybrid samples, the contact area of the junction between the inorganic and organic component is much enhanced, underlining the practical device utility of the concept presented in this paper. Vibrational spectroscopic measurements were performed to confirm the identity of the photoelectrodeposited polymers in the hybrid configurations. A Raman spectrum of the TiO2 NTA array indicates that TiO2 is in the anatase form, which is in agreement with previous reports (Figure 10).18,70 As for the hybrid sample, the typical vibrations of PANI are superimposed on the vibrations related to the TiO2 support. Assignment of the bands is given in the Supporting Information. Infrared spectra gave similar conclusions, namely, that the spectral pattern mirrors that measured for electrochemically synthesized polymer on a Au/Pt electrode. Finally, no indication was seen for overoxidation on any of the spectra, similar to the WO3 case.

Article

ASSOCIATED CONTENT

S Supporting Information *

Assignment of the bands in Raman and FT-IR spectra is given. The electroactivity of PANI in the hybrid configuration is shown. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (C.J.); [email protected] (K.R.), Tel: 817 272 5421. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS C.J. gratefully acknowledges the support of the European Union under FP7-PEOPLE-2010-IOF, grant number: 274046. We thank the colleagues at RenderNet, Ltd. for their support in the preparation of the artwork of the manuscript.

■

ABBREVIATIONS PPy, polypyrrole; PANI, polyaniline; CP, conducting polymer; ATR FT-IR, attenuated total reflection Fourier-transformed infrared spectroscopy; SEM, scanning electron microscopy; CV, cyclic voltammetry; SHE, standard hydrogen electrode

■

■

CONCLUDING REMARKS In this study, proof of concept was presented for photoelectrodeposition of conjugated polymers on nanostructured matrices of wide bandgap inorganic oxide semiconductors. Photocatalytic deposition preceding the electropolymerization step during potentiodynamic synthesis was shown to play a key role in the formation of homogeneous and organized assemblies of these hybrids. Employing illumination to deposit photoelectrochemically a very thin seed layer on the surface of an oxide semiconductor helps to overcome the main challenge, namely, its limited electroactivity. By varying the applied potential window during the potentiodynamic polymerization, the contribution of photoelectrochemical and electrochemical polymerization to the overall process could be separately expressed. Morphological features (as seen by FE-SEM) corroborated the decisive role of illumination in achieving homogeneous deposition of PANI and PPy on nanoporous WO3 and PANI on nanotubular TiO2. This uniform, skin-like polymer coating results in a much larger area of the polymer/ oxide semiconductor p/n junction, which is a prerequisite for most of the considered applications for these materials. Spectroscopic (FT-IR and Raman) studies not only corroborated the molecular identity of both components of the assemblies but also furnished important proof that there was insignificant overoxidation of the polymers during the synthetic procedure. This latter feature was also confirmed by cyclic voltammetric studies, which revealed the typical, quasireversible electroactivity for both polymers in the hybrid configuration. Finally, it is believed that bringing the inorganic nanostructures and the organic polymer into intimate contact on both physical and electronic levels (as demonstrated in this study) will facilitate practical applications of these materials, for example, in solar energy conversion.

REFERENCES

(1) Gomez-Romero, P. Adv. Mater. 2001, 13, 163. (2) Gangopadhyay, R.; De, A. Chem. Mater. 2000, 12, 608. (3) Nozik, A. J. Physica E 2002, 14, 115. (4) Rajeshwar, K.; de Tacconi, N. R.; Chenthamarakshan, C. R. Chem. Mater. 2001, 12, 2765. (5) Moon, S. J.; Baranoff, E.; Zakeeruddin, S. M.; Yeh, C.-Y.; Diau, E. W.-G.; Grätzel, M.; Sivula, K. Chem. Commun. 2011, 47, 8244. (6) Beek, W. J. E.; Wienk, M. M.; Janssen, R. A. J. Adv. Funct. Mater. 2006, 16, 1112−1116. (7) Bouclé, J.; Ravirajan, P.; Nelson, J. J. Mater. Chem. 2007, 17, 3141. (8) Coakley, K.; McGehee, M. D. Chem. Mater. 2004, 16, 4533. (9) Zhang, J.; Tu, J. P.; Zhang, D.; Qiao, Y. Q.; Xia, X. H; Wang, X.; Gu, C. D. J. Mater. Chem. 2011, 21, 17316. (10) Zhang, H.; Zong, R.; Zhu, Y. J. Phys. Chem. C 2009, 113, 4605. (11) Kerr, T.; Wu, H.; Nazar, L. Chem. Mater. 1996, 8, 2005. (12) Janáky, C.; Endrő di, B.; Berkesi, O.; Visy, C. J. Phys. Chem. C 2010, 114, 19338. (13) Zang, J.; Bao, S. J.; Li, C. M.; Bian, H.; Cui, X.; Bao, Q.; Sun, C. Q.; Guo, J.; Lian, K. J. Phys. Chem. C 2008, 112, 14843. (14) Huguenin, F.; Torresi, R. M. J. Phys. Chem. C 2008, 112, 2202. (15) Liu, R.; Lee, S. B. J. Am. Chem. Soc. 2008, 130, 2942. (16) Li, S. S.; Chang, C. P.; Lin, C. C.; Lin, Y. Y.; Chang, C. H.; Yang, J. R.; Chu, M. W.; Chen, C. W. J. Am. Chem. Soc. 2011, 133, 11614. (17) Liu, R.; Duay, J.; Lee, S. B. Chem. Commun. 2011, 47, 1384. (18) Chanmanee, W.; Watcharenwong, A.; Chenthamarakshan, C. R.; Kajitvichyanukul, P.; de Tacconi, N. R.; Rajeshwar, K. J. Am. Chem. Soc. 2008, 130, 965. (19) Roy, P.; Berger, S.; Schmuki, P. Angew. Chem., Int. Ed. 2011, 50, 2904. (20) Tsuchiya, H.; Macak, J. M.; Sieber, I.; Taveira, L.; Ghicov, A.; Sirotna, K.; Schmuki, P. Electrochem. Commun. 2005, 7, 295. (21) de Tacconi, N. R.; Chenthamarakshan, C. R.; Yogeeswaran, G.; Watcharenwong, A.; de Zoysa, R. S.; Basit, N. A; Rajeshwar, K. J. Phys. Chem. B 2006, 110, 25347. (22) Jennings, J. R.; Ghicov, A.; Peter, L. M.; Schmuki, P.; Walker, A. B. J. Am. Chem. Soc. 2008, 130, 13364. 19154



Article

(58) Debarros, R.; Deazevedo, W.; Deaguiar, F. Mater. Charact. 2003, 50, 131. (59) Quarto, F. D.; Figà, V.; Bocchetta, P.; Santamaria, M. Electrochem. Solid State 2007, 10, H305. (60) Di Franco, F.; Bocchetta, P.; Santamaria, M.; Di Quarto, F. Electrochim. Acta 2010, 56, 737. (61) Sapurina, I.; Riede, A.; Stejskal, J. Synth. Met. 2001, 123, 503. (62) Mu, S.; Kan, J. Electrochim. Acta 1996, 41, 1593. (63) Yang, H. J.; Bard, A. J. J. Electroanal. Chem. 1992, 339, 423. (64) Monllor-Satoca, D.; Borja, L.; Rodes, A.; Gómez, R.; Salvador, P. ChemPhysChem 2006, 7, 2540. (65) Lapkowski, M.; Berrada, K.; Quillard, S.; Louarn, G.; Lefrant, S.; Pron, A. Macromolecules 1995, 28, 1233. (66) Quillard, S.; Louran, G.; Lefrant, S.; MacDiarmid, A. G. Phys. Rev. B 1994, 50, 496. (67) Davidson, R. G.; Turner, T. G. Synth. Met. 1995, 72, 121. (68) Duchet, J.; Legras, R.; Demoustier-Champagne, S. Synth. Met. 1998, 95, 113. (69) Ghosh, S.; Bowmaker, G. A.; Cooney, R. P.; Seakins, J. M. Synth. Met. 1998, 95, 63. (70) Hardcastle, F. D.; Ishihara, H.; Sharma, R.; Biris, A. S. J. Mater. Chem. 2011, 21, 6337.

(23) Sun, W. T.; Yu, Y.; Pan, H. Y.; Gao, X. F.; Chen, Q.; Peng, L. M. J. Am. Chem. Soc. 2008, 130, 1124. (24) Baker, D. R.; Kamat, P. V. Adv. Funct. Mater. 2009, 19, 805. (25) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2006, 6, 215. (26) Song, Y. Y.; Schmidt-Stein, F.; Bauer, S.; Schmuki, P. J. Am. Chem. Soc. 2009, 131, 4230. (27) Zheng, H.; Ou, J. Z.; Strano, M. S.; Kaner, R. B.; Mitchell, A.; Kalantarzadeh, K. Adv. Funct. Mater. 2011, 21, 2175. (28) Nah, Y. C.; Ghicov, A.; Kim, D.; Schmuki, P. Electrochem. Commun. 2008, 10, 1777. (29) Kukkola, J.; Mäklin, J.; Halonen, N.; Kyllönen, T.; Tóth, G.; Szabó, M.; Shchukarev, A.; Mikkola, J.-P.; Jantunen, H.; Kordás, K. Sens. Actuators, B 2011, 153, 293. (30) Liu, J.; Kadnikova, E. N.; Liu, Y.; McGehee, M. D.; Fréchet, J. M. J. J. Am. Chem. Soc. 2004, 126, 9486. (31) Neyshtadt, S.; Jahnke, J. P.; Messinger, R. J.; Rawal, A; Segal Peretz, T.; Huppert, D.; Chmelka, B. F.; Frey, G. L. J. Am. Chem. Soc. 2011, 133, 10119. (32) Mor, G. K.; Kim, S.; Paulose, M.; Varghese, O. K.; Shankar, K.; Basham, J.; Grimes, C. A. Nano Lett. 2009, 9, 4250. (33) Shankar, K.; Mor, G. K.; Prakasam, H. E.; Varghese, O. K.; Grimes, C. A. Langmuir 2007, 23, 12445. (34) Lee, J.; Jho, J. Y. Sol. Energy Mater. Sol. Cells 2011, 95, 3152. (35) Atienzar, P.; Ishwara, T.; Horie, M.; Durrant, J. R.; Nelson, J. J. Mater. Chem. 2009, 19, 5377. (36) Koh, J. K.; Kim, J.; Kim, B.; Kim, J. H.; Kim, E. Adv. Mater. 2011, 23, 1641. (37) Tepavcevic, S.; Darling, S. B.; Dimitrijevic, N. M.; Rajh, T.; Sibener, S. J. Small 2009, 5, 1776. (38) Wang, D.; Liu, Y.; Wang, C.; Zhou, F.; Liu, W. ACS Nano 2009, 3, 1249. (39) Mujawar, S. H.; Ambade, S. B.; Battumur, T.; Ambade, R. B.; Lee, S.-H. Electrochim. Acta 2011, 56, 4462. (40) Janáky, C.; Bencsik, G.; Rácz, A.; Visy, C.; de Tacconi, N. R.; Chanmanee, W.; Rajeshwar, K. Langmuir 2010, 26, 13697. (41) Liang, H. C; Li, X. Z Appl. Catal. B-Environ. 2009, 86, 8. (42) Zhang, Z.; Yuan, Y.; Liang, L.; Cheng, Y.; Xu, H.; Shi, G.; Jin, L. Thin Solid Films 2008, 516, 8663. (43) Döbbelin, M.; Tena-Zaera, R.; Carrasco, P. M.; Sarasua, J.-R.; Cabañero, G.; Mecerreyes, D. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 4648. (44) Velazquez, J. M.; Gaikwad, A. V.; Rout, T. K.; Rzayev, J.; Banerjee, S. ACS Appl. Mater. Interfaces 2011, 3, 1238. (45) Janaky, C.; Tacconi, N. D.; Chanmanee, W.; Rajeshwar, K. J. Phys. Chem.C 2012, 116, 4234. (46) Kowalski, D.; Schmuki, P. Chem. Commun. 2010, 46, 8585. (47) Kowalski, D.; Tighineanu, A.; Schmuki, P. J. Mater. Chem. 2011, 17909. (48) Nielsch, K.; Müller, F.; Li, A. P.; Gösele, U. Adv. Mater. 2000, 12, 582. (49) Wang, D.; Ye, Q.; Yu, B.; Zhou, F. J. Mater. Chem. 2010, 20, 6910. (50) Chenthamarakshan, C. R.; Ming, Y.; Ming, Y. Chem. Mater. 2000, 12, 3538. (51) Somasundaram, S.; Chenthamarakshan, C. R.; de Tacconi, N. R.; Ming, Y.; Ming, Y; Rajeshwar, K. Chem. Mater. 2004, 16, 3846. (52) Fox, M. A.; Worthen, K. L. Chem. Mater. 1991, 3, 253. (53) Strandwitz, N. C.; Nonoguchi, Y.; Boettcher, S. W.; Stucky, G. D. Langmuir 2010, 26, 5319. (54) Okano, M.; Kikuchi, E.; Itoh, K.; Fujishima, A. J. Electrochem. Soc. 1988, 135, 1641. (55) de Tacconi, N. R.; Chenthamarakshan, C. R.; Rajeshwar, K.; Tacconi, E. J. J.Phys. Chem. B 2005, 109, 11953. (56) Kim, Y.; Lim, J. W.; Sung, Y. E.; Xia, J. B; Masaki, N.; Yanagida, S. J. Photochem. Photobiol., A 2009, 204, 110. (57) Breimer, M. A.; Yevgeny, G.; Sy, S.; Sadik, O. A. Nano Lett. 2001, 1, 305. 19155


Bringing Conjugated Polymers and Oxide ... - ACS Publications

Recommend Documents